Air chemistry sampling instruments will be installed in two portable buildings at a Barrow Alaska site during February and March 2009. The instruments will have inlet piping that extends a short distance outside of the building walls into the ambient air. A flow modeling study was performed to assess the likelihood that the building walls might contaminate the samples. This is especially important for measurements of reactive gases.
Wind climatology report
Portable buildings are 14'x20'x12' tall, supported on 12" high skids.
end photo |
side photo |
The portable buildings are modeled in Gambit as relatively simple arrangement of flat planes that represent the basic geometry. There are two buildings. Their long axes are at a 71° angle, and their closest separation is 14'. One building is north of the other. They are placed in a flow domain that is 200' along the x-axis (mean wind direction), 200' wide and 40' tall. In order to vary the direction of the ambient wind, the upwind boundary face is defined by an elliptical arc. The downwind boundary is a flat face. The computational grid is a tetrahedral-hybrid unstructured mesh, with an initial size of 0.2' on the building surfaces. The grid spacing expands exponentially to 1.0' when 5' from the buildings and reaches a maximum of 5' when 30' or further from the buildings. The domain has a total of 1,547,167 cells.
(click on images to see grid)
 |
 |
 |
The approach describe here produces useful results but fails to account for surface roughness, vertical shear at the domain edges, heat transfer or convection, variations in wind speed or direction. It does not account for static stability in the temperature profile that is typical for polar regions in winter. The top of the domain is modeled unrealistically as a rigid wall at 40', but is sufficiently far from the region of interest that it probably doesn't matter for the purposes of this study.